Adaptive Information Belief Space Planning

TL;DR

This paper introduces an efficient planning method for autonomous systems that explicitly reasons about uncertainty using an abstract observation model, reducing computational costs while maintaining decision quality.

Contribution

It proposes an approximation scheme with bounds on information-theoretic rewards and a refinement method to improve computational efficiency in belief space planning.

Findings

The abstract observation model reduces computational costs significantly.

Bounds on the expected reward guide the aggregation refinement.

The method achieves similar decision quality with less computation.

Abstract

Reasoning about uncertainty is vital in many real-life autonomous systems. However, current state-of-the-art planning algorithms cannot either reason about uncertainty explicitly, or do so with a high computational burden. Here, we focus on making informed decisions efficiently, using reward functions that explicitly deal with uncertainty. We formulate an approximation, namely an abstract observation model, that uses an aggregation scheme to alleviate computational costs. We derive bounds on the expected information-theoretic reward function and, as a consequence, on the value function. We then propose a method to refine aggregation to achieve identical action selection with a fraction of the computational time.